Fig. 3From: Adjustment for baseline characteristics in randomized trials using logistic regression: sample-based model versus true modelScatter plot of (sample based) adjusted estimates of the treatment effect (b1) as a function of the confounder imbalance between the two groups (Cohen’s d), with true parameter values β1 = 1 and β2 = − 1, for continuous and binary confounders, at sample size 2 × 50. Grey lines represent non-parametric regression functions (Lowess)Back to article page